You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
nate.river f75be5e264
fix t5 & llava (#1048)
1 week ago
..
_csrc/cuda update flash_attn_bwd (#1041) 2 weeks ago
_legacy update flash_attn_v2 (#1026) 2 weeks ago
abc add ia3 peft (#1033) 2 weeks ago
data add beit model & update pylint rules (#950) 1 month ago
dataset Trainer support 'map_fn' & add bloom finetune example (#1022) 2 weeks ago
diffusers support jetmoe & fix python id() caused bugs (#998) 3 weeks ago
engine fix map_fn error & ops.dropout caused training speed error (#1030) 2 weeks ago
modules add new Trainer like hf-transformers (#1000) 3 weeks ago
parallel fix map_fn error & ops.dropout caused training speed error (#1030) 2 weeks ago
peft support Adaption Prompt (#1042) 2 weeks ago
text2vec add text2vec module (#983) 1 month ago
transformers fix t5 & llava (#1048) 1 week ago
trl add new Trainer like hf-transformers (#1000) 3 weeks ago
utils add modelscope & wisemodel endpoint (#1045) 1 week ago
vocab add beit model & update pylint rules (#950) 1 month ago
workflow add beit model & update pylint rules (#950) 1 month ago
__init__.py add beit model & update pylint rules (#950) 1 month ago
amp.py add new amp module like torch.amp, usse autocast instead of amp level (#1036) 2 weeks ago
configs.py add modelscope & wisemodel endpoint (#1045) 1 week ago
injection.py fix t5 & llava (#1048) 1 week ago
readme.md add qwen2_moe & fix bugs (#965) 1 month ago